ABSTRACT This proposal is submitted in response to RFA-CA-17-025. Hepatocellular carcinoma (HCC) is the most common primary liver cancer and one of the few cancers with increasing incidence in the United States. Survival of patients with HCC is poor because most patients are diagnosed at late stages when treatment options are limited. HCC surveillance can detect cancers early and improve survival but it is inconsistently implemented. Not all liver nodules detected on surveillance imaging are malignant, these indeterminate nodules require follow up imaging to determine if they are HCC. This proposal assembles a team of specialists with complementary expertise in liver diseases, liver cancer, radiology, engineering, statistics, and information technology. The goals are to: (1) Improve HCC surveillance uptake in patients with cirrhosis by leveraging electronic medical records to remind healthcare providers and patients when they are due for screening tests; (2) Leverage analytic morphomics and machine learning to increase the detectability of HCC at an early stage; and (3) Increase detectability of HCC at an early stage using tagged HCC-specific peptides. We have worked closely together for many years and have participated/led many NIH funded clinical research networks. We have expertise and experience to carry out the proposed studies and will collaborate with other centers in the Liver Center Network to validate the methods/models we develop. We will also enroll patients with cirrhosis and collect data and specimens and provide access to existing well characterized data and specimen repositories that include both prevalent and incident HCC for Trans-Network studies. We are confident we will be able to contribute to this Liver Cancer Network and to early detection of HCC in multiple ways.